ZIYAO ZHANG
Hello!
I’m Ziyao, a creative technologist and innovation designer in New York City.
About Me
Email
CV
Home
Video Works
Generative Arts
Creative Hands-on Projects
Record wonderful and interesting moments in life.
Fun
Photography
📷☁️ AIBIC: AI-Based Behavior Interpretation for Customized Image Generation05/2024
Individual Project
#AI Creative Generation
#Behavior Analysis
#Non-verbal Behavior RecognitionIntroduction
Currently, the application of artificial intelligence in human-computer interaction is mostly limited to text and voice input. However, with the increasingly complex needs of users, traditional interaction methods have difficulty meeting the needs of the creative field. This project aims to break through these limitations and develop an interaction model based on non-verbal behaviors such as gestures, expressions, and emotions, so that AI can more naturally and accurately understand user intentions and improve the fluency and creativity of human-computer interaction.
Concept
How can AI be made to “understand you” better? This project uses a multimodal behavior capture system to enable artificial intelligence to recognize and understand users' natural behaviors (such as gestures and facial expressions), providing users with a more intuitive operating experience. Whether in artistic creation, design conception, or emotional expression, users can achieve efficient control of AI with simple gestures or facial expressions, achieving a seamless integration of creativity and technology.
Goals
- Multimodal interaction: Integrating gesture recognition, facial expression analysis and emotion detection technologies ensures that the user's natural behavior can be accurately captured by the system. This allows AI to rely not only on language, but also on non-verbal signals to understand complex creative needs.
- Real-time feedback and content optimization: Through the user's natural behavior input, AI can generate content in real time and dynamically optimize the generated content based on user behavior to meet complex aesthetic and creative needs.
- Automation of the creative process: Integrating AI and user interaction into the creative process reduces tedious repetitive work, allowing creative professionals to focus on core design ideas, improving efficiency and creative freedom.Prototype 1
Replace text with emoji/gesture prompt.Prototype 2 UserflowPrototype 2
How gestures/actions are translated into modifying the prompt and applied to image modification.
Application scenarios
Include but are not limited to:
- Design and creative industries: Designers and artists can collaborate with AI through gestures and expressions, making creative inspiration immediate and visual, shortening the creative process and improving efficiency.
- Accessibility technology: Provide users with physical disabilities or language barriers with an accessible way to interact with AI, easily operate the system through gestures and expressions, and enhance the inclusiveness of technology.
- Education and training: Teachers and students can use AI's instant feedback to interact through natural gestures and expressions to create a personalized learning and teaching experience.